Entity

Time filter

Source Type


Kanitz G.R.,BioRobotics Institute of the Scuola Superiore santAnna
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference | Year: 2011

In this paper we present surface electromyo-graphic (EMG) data collected from 16 channels on five unimpaired subjects and one transradial amputee performing 12 individual finger movements and a rest class. EMG were processed using a traditional Time Domain feature-set and classifiers: a Linear Discriminant Analysis (LDA) a k-Nearest Neighbors (k-NN) and Support Vector Machine (SVM). Using continuous datasets we show that it is possible to achieve an accuracy up to 80% across subjects. Thereafter possibilities to reduce the numbers of channels physically required, as well as the number of features have been investigated by means of a developed Genetic Algorithm (GA) that included a bonus system to reward eliminated features and channels. The classification was performed firstly on the full datasets and in later runs using the GA. The GA demonstrated high redundancy in the recorded 16 channel data as well as the insignificance of certain features. Although the GA optimization yielded to reduce 8 to 11 channels depending on the subject, such reduction had little to no effect on the classification accuracies. Source


Cipriani C.,BioRobotics Institute of the Scuola Superiore santAnna
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference | Year: 2011

In transradial amputees, the muscles in the residual forearm naturally employed by the unimpaired for flexing/extending the hand fingers, are the most appropriate targets, for multi-fingered prostheses control. However, once the prosthetic socket is manufactured and fitted on the residual forearm, the recorded EMG might not be originated only by the intention of performing finger movements, but also by the muscular activity needed to sustain the prosthesis itself. In this work, we preliminary show--on healthy subjects wearing a prosthetic socket emulator--that (i) variations in the weight of the prosthesis, and (ii) upper arm movements significantly influence the robustness of a traditional classifier based on k-nn algorithm. We show in simulated conditions that traditional pattern recognition systems do not allow the separation of the effects of the weight of the prosthesis because a surface recorded EMG pattern caused by the simple lifting or moving of the prosthesis is misclassified into a hand control movement. This suggests that a robust classifier should add to myoelectric signals, inertial transducers like multi-axes position, acceleration sensors or sensors able to monitor the interaction forces between the socket and the end-effector. Source


Pistohl T.,Northumbria University | Cipriani C.,BioRobotics Institute of the Scuola Superiore santAnna | Jackson A.,Northumbria University | Nazarpour K.,Northumbria University
Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS | Year: 2013

Powered hand prostheses with many degrees of freedom are moving from research into the market for prosthetics. In order to make use of the prostheses' full functionality, it is essential to find efficient ways to control their multiple actuators. Human subjects can rapidly learn to employ electromyographic (EMG) activity of several hand and arm muscles to control the position of a cursor on a computer screen, even if the muscle-cursor map contradicts directions in which the muscles would act naturally. We investigated whether a similar control scheme, using signals from four hand muscles, could be adopted for real-time operation of a dexterous robotic hand. Despite different mapping strategies, learning to control the robotic hand over time was surprisingly similar to the learning of two-dimensional cursor control. © 2013 IEEE. Source


Kanitz G.R.,BioRobotics Institute of the Scuola Superiore santAnna | Antfolk C.,Lund University | Cipriani C.,BioRobotics Institute of the Scuola Superiore santAnna | Sebelius F.,Lund University | Carrozza M.C.,BioRobotics Institute of the Scuola Superiore santAnna
Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS | Year: 2011

In this paper we present surface electromyo-graphic (EMG) data collected from 16 channels on five unimpaired subjects and one transradial amputee performing 12 individual finger movements and a rest class. EMG were processed using a traditional Time Domain feature-set and classifiers: a Linear Discriminant Analysis (LDA) a k-Nearest Neighbors (k-NN) and Support Vector Machine (SVM). Using continuous datasets we show that it is possible to achieve an accuracy up to 80% across subjects. Thereafter possibilities to reduce the numbers of channels physically required, as well as the number of features have been investigated by means of a developed Genetic Algorithm (GA) that included a bonus system to reward eliminated features and channels. The classification was performed firstly on the full datasets and in later runs using the GA. The GA demonstrated high redundancy in the recorded 16 channel data as well as the insignificance of certain features. Although the GA optimization yielded to reduce 8 to 11 channels depending on the subject, such reduction had little to no effect on the classification accuracies. © 2011 IEEE. Source


Cipriani C.,BioRobotics Institute of the Scuola Superiore santAnna | Dalonzo M.,BioRobotics Institute of the Scuola Superiore santAnna | Carrozza M.C.,BioRobotics Institute of the Scuola Superiore santAnna
IEEE Transactions on Biomedical Engineering | Year: 2012

A multisite, vibrotactile sensory substitution system, that could be used in conjunction with artificial touch sensors in multifingered prostheses, to deliver sensory feedback to upper limb amputees is presented. The system is based on a low cost/power/size smart architecture of off-the-shelf miniaturized vibration motors; the main novelty is that it is able to generate stimuli where both vibration amplitude and frequency as well as beat interference can be modulated. This paper is aimed at evaluating this system by investigating the capability of healthy volunteers to perceiveon their forearmsvibrations with different amplitudes and/or frequencies. In addition, the ability of subjects in spatially discriminating stimulations on three forearm sites and recognizing six different combinations of stimulations was also addressed. Results demonstrate that subjects were able to discriminate different force amplitudes exerted by the device (accuracies greater than 75); when both amplitude and frequency were simultaneously varied, the pure discrimination of amplitude/frequency variation was affected by the variation of the other. Subjects were also able to discriminate with an accuracy of 93 three different sites and with an accuracy of 78 six different stimulation patterns. © 2006 IEEE. Source

Discover hidden collaborations